Hi there, an answer from someone who's not an expert in the field but used to play around with time series: The action of simulating a process using with input parameters then estimating the parameters is not an invariant, especially when the process involves nonlinearities.
Did you try increasing simulation length and the burn-in to see if you get more stable results ? JC 2011/9/28 Bogey <a5462...@nepwk.com>: > Hi there, > > I'm trying to fit a GJR-GARCH Model using fGarch. I wanted to try that by > fitting an APARCH model with a fixed delta of 2 and a non-fixed gamma. So I > was simply trying to use: > > spec <- garchFit(~aparch(1,1),data=garchSim(),delta=2) > coef(spec) > > And sometimes, it's working like a charm and delta is indeed exactly 2 in > the resulting coefficient vector. > Frequently, though, the resulting delta coefficient has some other value, > usually somewhere between 0 and 2. > > Any ideas why? Am I doing something wrong or could this be a bug in fGarch? > > -- > View this message in context: > http://r.789695.n4.nabble.com/fGarch-Fitting-and-APARCH-Modell-with-fixed-delta-tp3850702p3850702.html > Sent from the R help mailing list archive at Nabble.com. > > ______________________________________________ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
